Courses – Fakultet tehničkih nauka u Novom Sadu

Subject: Computer vision and information extraction from multimedia data (17.IISD12)

Native organizations units: No data

Type of studies	Title
Doctoral Academic Studies	Mathematics in Engineering (Godina: 2, Winter)
Doctoral Academic Studies	Industrial Engineering / Engineering Management (Godina: 2, Winter)
Doctoral Academic Studies	Mathematics in Engineering (Godina: 1, Winter)
Doctoral Academic Studies	Information Systems Engineering (Godina: 2, Winter)

General information:

Category	Scientific-professional
Scientific or art field	Computer Engineering and Computer Communication
Interdisciplinary	No
ECTS	10

Educational goal:

The course is technology oriented and designed to provide an overview of the state-of-the-art in computer vision to the doctoral students, who need to have basic knowledge of information technology and image and video processing, mathematics or a related field. Upon completion of the course the students will gain theoretical knowledge and practical skills, which will allow them to apply the technology in question to analyze big visual data and embark on research projects in the area of computer vision and its applications in their primary research areas.

Educational outcome:

Students will obtain the knowledge and skills that enable them to effectively apply the techniques of using images and video, artificial intelligence and machine learning to extract information from multimedia content. They will be introduced to different problems in the domain computer vision and basic techniques used to solve them. Throughout the course they will be given a chance to take part in ongoing research projects, experiments and preparation of the results for publication. By the end of the course the students should have a draft of a scientific publication ready for submission to a relevant international scientific conference.

Course content:

The course will cover the following areas: techniques for coding and storing pictures and videos, image segmentation based on texture and color, object detection, classification, texture, detection of moving objects, tracking moving objects, detection of interesting behavior of objects and subjects. Theoretical classes will be complemented by hand-on training in the use of open source computer vision software to solve practical computer-vision problems.

Teaching methods:

Auditory lectures and laboratory training, supervised research work, seminar paper and an oral exam.

Literature:

Authors	Title	Year	Publisher	Language
Culibrk D., Mirkovic M., Zlokolica V., Pokric M., Crnojevic V., Kukolj D.	Salient Motion Features for Video Quality Assessment	2010	IEEE transactions on image processing	English
Paragios, N., Chen, Y., Faugeras, O.	Handbook of Mathematical Models in Computer Vision	2006	Springer, New York	English
Gary Bradski, Adrian Kaehler	Learning OpenCV: Computer Vision with the OpenCV Library	2008	O`Reilly Media	English
Gonzalez, R.C., Woods, R.E.	Digital Image Processing (3rd Edition)	2008	Prentice-Hall, Inc., Upper Saddle River	English
Culibrk, D., Marques, O., Socek, D., Kalva, H., Furht, B.	Neural Network Approach to Background Modeling for Video Object Segmentation	2007	IEEE Transactions on Neural Networks	English